{
  "id": "2102.06064",
  "version": "v1",
  "published": "2021-02-11T15:17:46.000Z",
  "updated": "2021-02-11T15:17:46.000Z",
  "title": "Uncertainty Propagation in Convolutional Neural Networks: Technical Report",
  "authors": [
    "Christos Tzelepis",
    "Ioannis Patras"
  ],
  "comment": "A PyTorch implementation is available under the MIT license here: https://github.com/chi0tzp/uacnn",
  "categories": [
    "cs.LG"
  ],
  "abstract": "In this technical report we study the problem of propagation of uncertainty (in terms of variances of given uni-variate normal random variables) through typical building blocks of a Convolutional Neural Network (CNN). These include layers that perform linear operations, such as 2D convolutions, fully-connected, and average pooling layers, as well as layers that act non-linearly on their input, such as the Rectified Linear Unit (ReLU). Finally, we discuss the sigmoid function, for which we give approximations of its first- and second-order moments, as well as the binary cross-entropy loss function, for which we approximate its expected value under normal random inputs.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2021-02-11T15:17:46.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "convolutional neural network",
      "technical report",
      "uncertainty propagation",
      "binary cross-entropy loss function",
      "uni-variate normal random variables"
    ],
    "tags": [
      "github project"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}